Data Analysis - Global Development Index

Dimensions

Following is the list of dimensions that were obtained as the result of intial pre-processing;

• Indicator
• Region
• Country
• Decade
• Year

Indicators

Indicators were further narrowed down by grouping them into logical categories. Following are the resulting handpicked indicators;

a. Social

• Pupil-teacher ratio, primary
• Government expenditure on education, total (% of government expenditure)
• Access to electricity (% of population)
• Birth rate, crude (per 1,000 people)
• Population growth (annual %)

b. Enviornmental

• CO2 emissions (metric tons per capita)
• Total greenhouse gas emissions (kt of CO2 equivalent)
• Forest area (sq. km)
• Population density (people per sq. km of land area)
• Life expectancy at birth, total (years)

c. Economic

• GDP per capita (current US$)
• GDP growth (annual %)
• GDP per capita growth (annual %)
• Expense (% of GDP)
• GNI per capita (constant 2010 US$)

Economic trends for regions and countries since 1960

p <- ggplot(datRegionalYearlyAvg, aes(year, gdp_capita, color = region)) + 
  geom_line() + 
  geom_segment(aes(xstart=1960, xend = 2018, yend = gdp_capita), linetype = 2, colour = 'grey') +
  geom_point(size = 2) + 
  geom_text(aes(x = 2018, label = region), hjust = 0) + 
  transition_reveal(year) + 
  coord_cartesian(clip = 'off') + 
  labs(title = 'National GDP', x='Year', y = 'GDP per Capits') + 
  theme_minimal() + 
  theme(plot.margin = margin(5.5, 40, 5.5, 5.5)) +
  ease_aes('linear', interval = 0.0001)

## Warning: Ignoring unknown aesthetics: xstart

animate(p, nframes = 300, fps=10)

Regional Economogy

• Middle east was the leader until 1990
• Scandinavia overtook North America in 1975 and 1995
• South Asia didn’t show any drastic growth
• Disparity started increasing after 1990

p <- ggplot(datNationalYearlyAvg, aes(year, gdp_capita, color = country)) + 
  geom_line() + 
  geom_segment(aes(xstart=1960, xend = 2018, yend = gdp_capita), linetype = 2, colour = 'grey') +
  geom_point(size = 2) + 
  geom_text(aes(x = 2018, label = country), hjust = 0) + 
  transition_reveal(year) + 
  coord_cartesian(clip = 'off') + 
  labs(title = 'National GDP', x='Year', y = 'GDP per Capits') + 
  theme_minimal() + 
  theme(plot.margin = margin(5.5, 40, 5.5, 5.5)) +
  ease_aes('linear', interval = 0.0001)

## Warning: Ignoring unknown aesthetics: xstart

animate(p, nframes = 300, fps=10)

International Economoy

• UAE and Qatar had highest gdp growths until 1970
• Norway and Denmark showed highest growth in recent decade

Effect of Social prosperity and enviornmental prosperity over economy, before and after 2000

getCorrelation <- function(df){
# slice important cols
  cols <- c(
    "social_prosperity",
    "enviornment_friendly",

    "gdp_growth", # GDP growth (annual %)
    "gni_capita", # GNI per capita (constant 2010 US$)
    # "gni_capita_constant", # GNI per capita (constant LCU)
    
    "education_expense", # Government expenditure on education, total (% of government expenditure)
    "electricity_access", # Access to electricity (% of population)
    "population_growth", # Population growth (annual %)
    "life_expectancy", # Life expectancy at birth, total (years)
    
    "co2_emission", # CO2 emissions (metric tons per capita)
    "forest_area", # Forest area (sq. km)
    "population_density" # Population density (people per sq. km of land area)
  )
  df.slice <- df[,cols]
  
  
  df.slice.corr <- rcorr(as.matrix(df.slice))
  
  return(df.slice.corr)
}

par(mfrow=c(1,2))

# plot 
  # pre-2000
  datGlobalYearlyAvg.slice.corr.1960 <- getCorrelation(datNationalYearlyAvg[datNationalYearlyAvg$year<2000,])
    corrplot(datGlobalYearlyAvg.slice.corr.1960$r, type="upper",
         p.mat = datGlobalYearlyAvg.slice.corr.1960$P, sig.level = 0.01, insig = "blank")
  # 
  # post-2000
  datGlobalYearlyAvg.slice.corr.2018 <- getCorrelation(datNationalYearlyAvg[datNationalYearlyAvg$yea>=2000,])
  corrplot(datGlobalYearlyAvg.slice.corr.2018$r, type="upper",
         p.mat = datGlobalYearlyAvg.slice.corr.2018$P, sig.level = 0.01, insig = "blank")

Before 2000

• Electricity access decreased, with increase in Population density
• Life Expectancy is heavily dependent on GNI and Social Prosperity

After 2000

• All dependencies are still relevant
• Strength of dependencies intensified

History Over Time: Country’s journey to prosperity/failure from 1960 to 2018

• We were curious to know what is the secret of the successful countries.
• The report shows the indicators that have the greatest impact on making countries “rich and healthy”.
• We would like to know if we can find the “formula” of the richest and most comfortable countries based on the facts obtained from the indicators.
• We took y axis for life expectancy and x axis for wealth - Gross National Income Of countries.
• We wanted to visualise the “movement” of the countries on the graph based on mentioned parameters above. From “poor and sick” up to “rich and healthy”.
• At first, we will show the world 58 years ago in 1960, over the countries of Africa, Asia, Middle East, America and Europe.
• The size of the point area shows the population density.

  economic_indicators <- c(
    "gdp_capita", # GDP per capita (current US$)
    "gdp_growth", # GDP growth (annual %)
    "gdp_capita_growth", # GDP per capita growth (annual %)
    "expense_gdp", # Expense (% of GDP)
    "net_income" # GNI per capita (constant LCU)
  )

  social_indicators <- c(
    "teacher_per_pupil", # Pupil-teacher ratio, primary
    "education_expense", # Government expenditure on education, total (% of government expenditure)
    "electricity_access", # Access to electricity (% of population)
    "birth_rate", # Birth rate, crude (per 1,000 people)
    "population_growth" # Population growth (annual %)
  )
  
    environmental_indicators <- c(
    "co2_emission", # CO2 emissions (metric tons per capita)
    "greenhouse_emission", # Total greenhouse gas emissions (kt of CO2 equivalent)
    "forest_area", # Forest area (sq. km)
    "population_density", # Population density (people per sq. km of land area)
    "life_expectancy" #Life expectancy at birth, total (years)
  )
    
read_data <- read.csv('wdi-f.csv')

test <- read_data[FALSE,]

for(index in 1960:2020){
    test <- rbind(test, read_data[read_data$year==index,])
}

gg <- ggplot(test, aes(net_income, life_expectancy, color = region)) +
  geom_point(aes(size = population_density, frame = year, ids = country)) +
  scale_x_log10()+ labs(x = "GNI per capita")

ggplotly(gg)

• Colombia / Chile / Costa Rica have a good leap towards higher income
• Benin has a big leap towards life expectancy
• Life expectancy in Bostwana decreased around 1990~2000
• The life expectancy of Ecuador (the poorest country) increases significantly
• Life expectancy in Bangladesh increased to 75 from 45 years over the period (1975-2018)

Regional population growth trends since 1960

# normalize
datRegionalYearlyAvg$population_growth <- normalize(datRegionalYearlyAvg$population_growth)

# extract region
region1 <- datRegionalYearlyAvg[datRegionalYearlyAvg$region=="North America",]
region2 <- datRegionalYearlyAvg[datRegionalYearlyAvg$region=="Scandinavia",]
region3 <- datRegionalYearlyAvg[datRegionalYearlyAvg$region=="South Asia",]
region4 <- datRegionalYearlyAvg[datRegionalYearlyAvg$region=="Middle East",]


# plot
plot(region1$year, region1$population_growth*100, ylim=c(0,100), type = "o", col = 100, pch=19, xlab="Year", ylab = "Growth %", main = "Regional Population Growth")
lines(region2$year, region2$population_growth*100, type = "o", col = 110, pch=20)
lines(region3$year, region3$population_growth*100, type = "o", col = 120, pch=21)
lines(region4$year, region4$population_growth*100, type = "o", col = 130, pch=22)

legend("topright",
  legend = c("North America", "Scandinavia", "South Asia", "Middle East"),
  col = c(100,110, 120,130),
  pch = c(19,20,21,22)
)

• Middle east was most populated in history
• North America overtook Scandinavia region in late 2020
• Trends didn’t change throughout history

Good seprators (indicators) for rich and poor countries

#Data has been merged before and saved it as a new csv for the sake of simplicity
grouped_comparison_wide <- read.csv('Group_comparison.csv')

grouped_comparison_long <- wide_to_long(grouped_comparison_wide)

grouped_comparison_long <- renameIndicators(grouped_comparison_long)

grouped_comparison_long<-grouped_comparison_long[!is.na(grouped_comparison_long$value),]

grouped_comparison_indicators_wide <- dcast(grouped_comparison_long, year+country ~ indicator, value.var="value")
#grouped_comparison_indicators_wide[, -c(1,2)] <- scale(grouped_comparison_indicators_wide[, -c(1,2)])


long<-melt(grouped_comparison_indicators_wide, id.vars=c("year","country"))

llong<-within(long, {
    group = ifelse(variable %in% social_indicators, "Social", "Environmental")
    group = ifelse(variable %in% environmental_indicators, "Environmental", "Economical")
})

check_val <- function(x) {
  #print(x)
  if(x%in% social_indicators) {
    "Social"
    #print("Social")
  }
  else if(x%in% environmental_indicators) {
    "Environmental"
    #print("Environmental")
  }
  else if(x%in% economic_indicators) {
    "Economical"
    #print("Economical")
  }
}

llong$group <- apply(array(long[["variable"]]),MARGIN=1, FUN=check_val)
llong <- llong[!is.na(llong$value),]
#llong<-llong[llong$year<-2008,]
llong$year <-as.numeric(as.character(llong[,1])) 

plot_circular_bar<-function(llong,var1,var2){
  
  # library
  library(tidyverse)
  llong<-llong[llong$group==var1,]
  llong<-llong[llong$country==var2,]
  #view(llong)
  llong$variable<-factor(llong$variable)
  #print( llong$variable) 
  # Create dataset
  if (var1=="Environmental")
  {
    for (i in 1:length (environmental_indicators)){
      #print(environmental_indicators[i])
      llong[llong$variable==environmental_indicators[i],]$value = normalize(llong[llong$variable==environmental_indicators[i],]$value)
      
    }
      data=data.frame(
        individual=llong$year,
        group=llong$variable ,
        value=llong$value*50)
  }
  else if (var1=="Economical"){
    
        for (i in 1:length (economic_indicators)){
      #print(environmental_indicators[i])
      llong[llong$variable==economic_indicators[i],]$value = normalize(llong[llong$variable==economic_indicators[i],]$value)
      
    }
    
      data=data.frame(
        individual=llong$year,
        group=llong$variable ,
        value=llong$value*50)
  }
  else{
            for (i in 1:length (social_indicators)){
      #print(environmental_indicators[i])
      llong[llong$variable==social_indicators[i],]$value = normalize(llong[llong$variable==social_indicators[i],]$value)
      
    }
    
      data=data.frame(
        individual=llong$year,
        group=llong$variable ,
        value=llong$value*50)
  }

  #data = data %>% arrange(group, value)    
  
  # Set a number of 'empty bar' to add at the end of each group
  empty_bar=2
  to_add = data.frame( matrix(NA, empty_bar*nlevels(data$group), ncol(data)) )
  colnames(to_add) = colnames(data)
  to_add$group=rep(levels(data$group), each=empty_bar)
  data=rbind(data, to_add)
  data=data %>% arrange(group)
  data$id=seq(1, nrow(data))
  
  # Get the name and the y position of each label
  label_data=data
  number_of_bar=nrow(label_data)
  angle= 90 - 360 * (label_data$id-0.5) /number_of_bar     # I substract 0.5 because the letter must have the angle of the center of the bars. Not extreme right(1) or extreme left (0)
  label_data$hjust<-ifelse( angle < -90, 1, 0)
  label_data$angle<-ifelse(angle < -90, angle+180, angle)
  
  # prepare a data frame for base lines
  base_data=data %>% 
    group_by(group) %>% 
    summarize(start=min(id), end=max(id) - empty_bar) %>% 
    rowwise() %>% 
    mutate(title=mean(c(start, end)))
  
  # prepare a data frame for grid (scales)
  grid_data = base_data
  grid_data$end = grid_data$end[ c( nrow(grid_data), 1:nrow(grid_data)-1)] + 1
  grid_data$start = grid_data$start - 1
  grid_data=grid_data[-1,]
  
  
  # Make the plot
  p = ggplot(data, aes(x=as.factor(id), y=value, fill=group)) +       # Note that id is a factor. If x is numeric, there is some space between the first bar
    
    geom_bar(aes(x=as.factor(id), y=value, fill=group), stat="identity", alpha=0.5) +
    
    # Add a val=100/75/50/25 lines. I do it at the beginning to make sur barplots are OVER it.
    geom_segment(data=grid_data, aes(x = end, y = 80, xend = start, yend = 80), colour = "grey", alpha=1, size=0.3 , inherit.aes = FALSE ) +
    geom_segment(data=grid_data, aes(x = end, y = 60, xend = start, yend = 60), colour = "grey", alpha=1, size=0.3 , inherit.aes = FALSE ) +
    geom_segment(data=grid_data, aes(x = end, y = 40, xend = start, yend = 40), colour = "grey", alpha=1, size=0.3 , inherit.aes = FALSE ) +
    geom_segment(data=grid_data, aes(x = end, y = 20, xend = start, yend = 20), colour = "grey", alpha=1, size=0.3 , inherit.aes = FALSE ) +
    
    # Add text showing the value of each 100/75/50/25 lines
    annotate("text", x = rep(max(data$id),4), y = c(20, 40, 60, 80), label = c("20", "40", "60", "80") , color="grey", size=1 , angle=0, fontface="bold", hjust=1) +
    
    geom_bar(aes(x=as.factor(id), y=value, fill=group), stat="identity", alpha=0.5) +
    ylim(-100,120) +
    theme_minimal() +
    theme(
      legend.position = "right",
      axis.title = element_blank(),
      axis.text = element_blank(),
      panel.grid = element_blank(),
      plot.margin = unit(rep(-1,4), "cm") 
    ) +
    coord_polar() + 
    geom_text(data=label_data, aes(x=id, y=value+60, label=individual, hjust=hjust), color="black",alpha=0.6, size=2.5, angle= label_data$angle, inherit.aes = FALSE ) +
    
    # Add base line information
    geom_segment(data=base_data, aes(x = start, y = -5, xend = end, yend = -5), colour = "black", alpha=0.8, size=0.6 , inherit.aes = FALSE )
  
  #   if (var1=="Economical")
  #   {
  #       #print(var1)
  #       #print(var2)
  #       if(var2=="Low income"){
  #         geom_text(data=base_data, aes(x = title, y = -18, label=group), hjust=c(1,1,0), colour = "black", alpha=0.8, size=2, fontface="bold", inherit.aes = FALSE)
  #       }
  #         else
  #       {
  #                   geom_text(data=base_data, aes(x = title, y = -18, label=group), hjust=c(1,1,1,0), colour = "black", alpha=0.8, size=2, fontface="bold", inherit.aes = FALSE)
  #       }
  #   }
  # else
  # {
  #         #print(base_data)
  #         geom_text(data=base_data[1:5,], aes(x = title, y = -18, label=group), hjust=c(1,1,1,0,1), colour = "black", alpha=0.8, size=3, fontface="bold", inherit.aes = FALSE)
  # }
 if(var1=="Economical" && var2=="Low income")
 {
   #print(paste(var2,":",var1))
   #"#9ACD32", "#00CED1", "#BA55D3"
   p + scale_fill_manual(paste(var2,":",var1), values = c("gdp_growth" = "#9ACD32", "gdp_capita" = "#00CED1", "gdp_capita_growth" = "#BA55D3"))
 }
  else{
     p + scale_fill_discrete(name = paste(var2,":",var1))
  }
  
  
  #print(data$group)
}

plot_circular_bar(llong,"Social","High income")

plot_circular_bar(llong,"Social","Middle income")

plot_circular_bar(llong,"Social","Low income")

plot_circular_bar(llong,"Environmental","High income")

plot_circular_bar(llong,"Environmental","Middle income")

plot_circular_bar(llong,"Environmental","Low income")

plot_circular_bar(llong,"Economical","High income")

plot_circular_bar(llong,"Economical","Middle income")

plot_circular_bar(llong,"Economical","Low income")

Effect of social prosperity on economic prosperity

ggplot(datRegionalDecadeAvg, aes(decade, social_prosperity, fill = social_prosperity)) +
  geom_col() +
  scale_fill_distiller(palette = "Reds", direction = 1) +
  facet_wrap(~region) +
  labs(x = 'Decade', y = 'Social Prosperity Score') +
  transition_states(decade) +
  shadow_mark()

• Economic success resulted social prosperity for region North America and Scandinavia
• Social prosperity of region South Asia and Middle East has not witness any drastic change